matthew green | 7 Aug 10:32 2011
Picon

HEADS UP: port i386 and amd64 switched to GCC 4.5.3


hi folks.

i've switched i386 and amd64 to use GCC 4.5.3 by default.  there
shouldn't be any major issues, however update builds won't work.
i recommend removing both destdir and objdir before rebuilding.

if you need to revert this change, use build.sh -V HAVE_GCC=4.

let me know if there are problems!  thanks.

.mrg.

Cherry G. Mathew | 7 Aug 16:56 2011
Picon

merging bits of [cherry-xenmp]

Hi,

I'd like to start merging in bits of the cherry-xenmp branch into
-current over the coming week. The changes should be transparent, and
shouldn't change any behaviour of -current. These include a few cleanups
of MD code and some MP related changes (eg: to x86 pmap)  that should be
agnostic to XEN UP, port-i386 and port-amd64. Where relevant, I will
make sure that code is reviewed before committing.

Please let me know in the next couple of days if there are serious
objections to this.

Cheers,

--

-- 
Cherry

PS: Xen MP support is not intended through these merges.

Chavdar Ivanov | 8 Aug 01:37 2011
Picon

Re: HEADS UP: port i386 and amd64 switched to GCC 4.5.3

On 7 August 2011 09:32, matthew green <mrg <at> eterna.com.au> wrote:
>
> hi folks.
>
>
> i've switched i386 and amd64 to use GCC 4.5.3 by default.  there
> shouldn't be any major issues, however update builds won't work.
> i recommend removing both destdir and objdir before rebuilding.
>
> if you need to revert this change, use build.sh -V HAVE_GCC=4.
>
> let me know if there are problems!  thanks.

Nice; no problems so far as far as I can see:

[uksup2] ~ $ /usr/bin/gcc --version
gcc (NetBSD nb1 20110620) 4.5.3
Copyright (C) 2010 Free Software Foundation, Inc.
This is free software; see the source for copying conditions.  There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.

[uksup2] ~ $ uname -a
NetBSD uksup2 5.99.55 NetBSD 5.99.55 (GENERIC) #0: Sun Aug  7 15:18:14
BST 2011  root <at> uksup2:/usr/obj/usr-current/i386/sys/arch/i386/compile/GENERIC
i386

I didn't even notice it...

>
>
(Continue reading)

Hubert Feyrer | 8 Aug 01:44 2011
Picon

Re: HEADS UP: port i386 and amd64 switched to GCC 4.5.3


>> let me know if there are problems!  thanks.

Cross-building from OS X, I get:

#   compile  ramdisk-g4u/getcap.o
/Users/feyrer/work/NetBSD/cvs/src-g4u/obj.i386/tooldir/bin/i486--netbsdelf-gcc 
-Os -std=gnu99 -Werror 
--sysroot=/Users/feyrer/work/NetBSD/cvs/src-g4u/obj.i386/destdir  -DSMALL 
-c

/Users/feyrer/work/NetBSD/cvs/src-g4u/distrib/utils/libhack/../../../lib/libc/gen/getcap.c
/Users/feyrer/work/NetBSD/cvs/src-g4u/distrib/utils/libhack/../../../lib/libc/gen/getcap.c:1213:1: 
internal compiler error: in execute_ipa_pass_list, at passes.c:1800
Please submit a full bug report,
with preprocessed source if appropriate.
See <http://www.NetBSD.org/Misc/send-pr.html> for instructions.

*** Failed target:  getcap.o
*** Failed command: 
/Users/feyrer/work/NetBSD/cvs/src-g4u/obj.i386/tooldir/bin/i486--netbsdelf-gcc 
-Os -std=gnu99 -Werror 
--sysroot=/Users/feyrer/work/NetBSD/cvs/src-g4u/obj.i386/destdir -DSMALL 
-c 
/Users/feyrer/work/NetBSD/cvs/src-g4u/distrib/utils/libhack/../../../lib/libc/gen/getcap.c
*** Error code 1

  - Hubert
matthew green | 8 Aug 01:47 2011
Picon

re: HEADS UP: port i386 and amd64 switched to GCC 4.5.3


> Cross-building from OS X, I get:
> 
> #   compile  ramdisk-g4u/getcap.o
> /Users/feyrer/work/NetBSD/cvs/src-g4u/obj.i386/tooldir/bin/i486--netbsdelf-gcc 
> -Os -std=gnu99 -Werror 
> --sysroot=/Users/feyrer/work/NetBSD/cvs/src-g4u/obj.i386/destdir  -DSMALL 
> -c 
> /Users/feyrer/work/NetBSD/cvs/src-g4u/distrib/utils/libhack/../../../lib/libc/gen/getcap.c
>
/Users/feyrer/work/NetBSD/cvs/src-g4u/distrib/utils/libhack/../../../lib/libc/gen/getcap.c:1213:1: 
> internal compiler error: in execute_ipa_pass_list, at passes.c:1800
> Please submit a full bug report,
> with preprocessed source if appropriate.
> See <http://www.NetBSD.org/Misc/send-pr.html> for instructions.
> 
> *** Failed target:  getcap.o
> *** Failed command: 
> /Users/feyrer/work/NetBSD/cvs/src-g4u/obj.i386/tooldir/bin/i486--netbsdelf-gcc 
> -Os -std=gnu99 -Werror 
> --sysroot=/Users/feyrer/work/NetBSD/cvs/src-g4u/obj.i386/destdir -DSMALL 
> -c 
> /Users/feyrer/work/NetBSD/cvs/src-g4u/distrib/utils/libhack/../../../lib/libc/gen/getcap.c
> *** Error code 1

what build.sh command and environment are you using?
did you try a totally clean build?

thanks,

(Continue reading)

Hugo Silva | 8 Aug 09:55 2011

Re: merging bits of [cherry-xenmp]

On 08/07/11 15:56, Cherry G. Mathew wrote:
> Hi,
> 
> I'd like to start merging in bits of the cherry-xenmp branch into
> -current over the coming week. The changes should be transparent, and
> shouldn't change any behaviour of -current. These include a few cleanups
> of MD code and some MP related changes (eg: to x86 pmap)  that should be
> agnostic to XEN UP, port-i386 and port-amd64. Where relevant, I will
> make sure that code is reviewed before committing.
> 
> 
> Please let me know in the next couple of days if there are serious
> objections to this.
> 
> Cheers,
> 

Hello Cherry,

From a user/sysadmin perspective, will there be any changes? Or is this
"beach front" work for more changes to come?

In either case I'll be happy to build a more recent -current with your
changes in and see if all works as expected, if this helps.

Regards,

Hugo

(Continue reading)

Hubert Feyrer | 8 Aug 10:04 2011
Picon

re: HEADS UP: port i386 and amd64 switched to GCC 4.5.3

On Mon, 8 Aug 2011, matthew green wrote:
> what build.sh command and environment are you using?
> did you try a totally clean build?

build command is:

 	arch=i386
 	time /usr/bin/nice -n 3 \
 	                        sh build.sh \
 	                                -N 2 \
 	                                -U \
 	                                -m i386 \
 					-u -T obj.${arch}/tooldir -D obj.${arch}/destdir -R obj.${arch}/releasedir
 		                        tools \
 	                                distribution \
 	                                kernel=G4U-KERNEL

environment is:

 	PATH=/usr/local/texlive-2010/bin/universal-darwin:/Volumes/pkgsrc5/pkg/bin:/Volumes/pkgsrc5/pkg/sbin:/usr/bin:/bin:/usr/sbin:/sbin:/usr/local/bin:/usr/local/MacGPG2/bin:/usr/X11/bin:/bin:/usr/bin:/usr/sbin:/sbin:/usr/ucb:/Users/feyrer/bin:/usr/pkg/bin:/usr/pkg/sbin:/usr/ccs/bin:/usr/games:/usr/X11R6/bin:/usr/X11R6/sbin:/usr/X11R7/bin:/usr/X11R7/sbin:/usr/local/bin
 	TMPDIR=/var/folders/p3/p3hc1hhhEz0eJ8oK4LJRKE+++TM/-Tmp-/
 	SHELL=/bin/tcsh
 	HOME=/Users/feyrer
 	USER=feyrer
 	LOGNAME=feyrer
 	GPG_AGENT_INFO=/Users/feyrer/.gnupg/S.gpg-agent:150:1
 	DISPLAY=:0.0
 	SSH_AUTH_SOCK=/tmp/launch-dwtKe3/Listeners
 	Apple_PubSub_Socket_Render=/tmp/launch-RBJsVb/Render
 	COMMAND_MODE=legacy
(Continue reading)

haad | 8 Aug 12:44 2011
Picon

Re: merging bits of [cherry-xenmp]

Hi,

On Sun, Aug 7, 2011 at 4:56 PM, Cherry G. Mathew <cherry <at> zyx.in> wrote:
> Hi,
>
> I'd like to start merging in bits of the cherry-xenmp branch into
> -current over the coming week. The changes should be transparent, and
> shouldn't change any behaviour of -current. These include a few cleanups
> of MD code and some MP related changes (eg: to x86 pmap)  that should be
> agnostic to XEN UP, port-i386 and port-amd64. Where relevant, I will
> make sure that code is reviewed before committing.

Can you share a diff with us ?

>
>
> Please let me know in the next couple of days if there are serious
> objections to this.
>
> Cheers,
>
> --
> Cherry
>
> PS: Xen MP support is not intended through these merges.
>

--

-- 

Regards.
(Continue reading)

Cherry G. Mathew | 9 Aug 08:01 2011
Picon

Re: merging bits of [cherry-xenmp]

>>>>> "haad" == haad  <haaaad <at> gmail.com> writes:

    haad> Hi,
    haad> On Sun, Aug 7, 2011 at 4:56 PM, Cherry G. Mathew <cherry <at> zyx.in> wrote:
    >> Hi,
    >> 
    >> I'd like to start merging in bits of the cherry-xenmp branch into
    >> -current over the coming week. The changes should be transparent,
    >> and shouldn't change any behaviour of -current. These include a
    >> few cleanups of MD code and some MP related changes (eg: to x86
    >> pmap)  that should be agnostic to XEN UP, port-i386 and
    >> port-amd64. Where relevant, I will make sure that code is
    >> reviewed before committing.

    haad> Can you share a diff with us ?

Hi Adam - please have a look at the [cherry-xenmp] branch for details. 

Cheers,
--

-- 
Cherry

Cherry G. Mathew | 10 Aug 01:37 2011
Picon

Re: merging bits of [cherry-xenmp]

>>>>> "haad" == haad  <haaaad <at> gmail.com> writes:

    haad> Hi,
    haad> On Sun, Aug 7, 2011 at 4:56 PM, Cherry G. Mathew <cherry <at> zyx.in> wrote:
    >> Hi,
    >> 
    >> I'd like to start merging in bits of the cherry-xenmp branch into
    >> -current over the coming week. The changes should be transparent,
    >> and shouldn't change any behaviour of -current. These include a
    >> few cleanups of MD code and some MP related changes (eg: to x86
    >> pmap)  that should be agnostic to XEN UP, port-i386 and
    >> port-amd64. Where relevant, I will make sure that code is
    >> reviewed before committing.

    haad> Can you share a diff with us ?

Here's a sample:
ftp://ftp.netbsd.org/pub/NetBSD/misc/cherry/tmp/port-xen/breakout1/earlyglue1.diff

comments ?

--

-- 
Cherry

PS: tech-kern <at>  is off Cc:


Gmane